In a report released on its website, ISE is describing how it was able to take control of an iPhone by getting malicious code installed by three methods.
The first method is the scariest as it deals with the iPhone’s learning which Wi-Fi networks to trust. If you have picked a trusted network, and find an attacker controlled access point of the same name and encryption, the code will be auto loaded on to your phone. The odds of this seem fairly low, but are still frightening in that there is so little user intervention.
The other two methods involve visiting an online forum with the code embedded in a message, and the third one involves opening a link included in an email or SMS message.
All three methods of course can be avoided by applying normal safe-surfing procedures. Use only networks you trust, visit sites you know and trust, and don’t open links from unknown senders.
The vulnerability does also exist in Windows and Mac versions of the Apple produced web browser. At this time, this has only been accomplished in a laboratory setting with no reports of a real-world occurrence. Apple has been notified of the vulnerability and is looking at it.
There is a preliminary version of their paper on their site, and a final version will be released at BlackHat in Las Vegas on August 2nd.